Chapter 12 Inferring from the Data Inferring from

Chapter 12 Inferring from the Data

Inferring from Data Estimation and Significance testing

Inferential Statistics Set of statistical procedures that allow us to go beyond the group measured to make statements about the characteristics of a larger group. n Estimation n Significance testing n

Inferential Statistics n Estimation - used to generalize the results from sample to population n Significance testing - examines how likely differences between groups and relationships between variables occur by chance

Estimation n Most populations are too large to conduct a census …. . So we sample n Statistics allow us to “estimate” characteristics back to the population n Example: % of Native Americans in Cleveland

Only possible when we meet 2 assumptions n Variables of interest are assumed to have a normal distribution n use of a random sample

Assumption #1 n Estimation begins with the assumption that the scores for a particular variable are distributed in the shape of a symmetrical bell (Bell curve) n This is called normal distribution

Assumption #2 n Random sampling-- that each member of the population has an equal chance to be included n when inferring back to population… n how confident are you when estimating population parameters ? 95% is the accepted level in social sciences

Significance Testing n When trying to answer questions about how variables are related to each other we test for the significance level

Significance Testing Con’t An implicit assumption is built into each hypothesis n It predicts that there is no difference between groups and/or no relationship between the variables n represents chance occurrence n NULL HYPOTHESIS n

Significance Testing Con’t n Significance testing--the process of analyzing quantitative data for the purpose of testing whether a null hypothesis is either correct or false n We want to reject the null hypothesis

Significance Testing Con’t n Significance level -- the probability level that establishes that we have enough confidence to reject the null hypothesis n A 95% or. 05 level is standard